We Can't Let Enormous Weirdos Regulate AI
6 hours ago
- The US approach to AI regulation is criticized for focusing on bizarre existential fantasies (like 'Superintelligent Thumb Dogs with Drones') rather than tangible, real-world harms such as scams, job loss, and AI-powered surveillance.
- Two main factions are identified: the 'AI Doom Contingent' (Anthropic, OpenAI, EA) pushing for a regulatory slowdown/cartel, and the 'Dead-Eyed Capitalists' (Andreessen, Zuckerberg, Huang) pushing for unhindered acceleration.
- A central critique is the incestuous social, financial, and ideological overlap between AI company executives, 'safety' researchers, and the Rationalist/Effective Altruist communities in the Bay Area, undermining objective regulation.
- Effective Altruism and the AI doomer movement are accused of siphoning funding and political attention away from practical harms and experienced policy experts, effectively taking over the country's tech regulatory space.
- The author calls for breaking the insular circle by educating the public, winning political power, and empowering demographically diverse, level-headed experts with relevant experience to regulate AI sensibly.